A previously funded study recently completed the intake of 98 patients with nonmanic functional psychosis, each rigorously diagnosed according to three systems (RDC, DSM-III and Feighner et al). Baseline assessment included neuroendocrine measures as well as detailed, clinical and demographic descriptions. A preliminary analysis of the six and 12 month follow-up incorporated in that study has led to the following proposal that follow-up be extended to three and five years. This extension is intended to accomplish several goals. First, it will determine the eventual likelihood of recovery for the many patients who remain psychotic at one year and will quantify relapse rates and the risk of eventual chronicity within groups defined alternatively by diagnostic and neuroendocrine measures at baseline. Second, it will discover time sensitive outcome predictors not apparent at 12 months. Preliminary analysis has revealed such a predictor in that the dexamethasone suppression test strongly predicted outcome at 12 months (independent of diagnosis) but not at six months. Third, the extended follow-up will permit a description of major diagnostic evolution, a phenomenon particularly likely in these patient groups. Finally, the lengthier follow-up and the measures incorporated in it will facilitate the comparisons with other relevant past and ongoing studies which use five years as an end point. To accomplish this, a highly trained rater will interview each subject at three years and five years using an instrument designed to assess proximal two year, one year and one month periods in progressively greater detail, yielding a longitudinal record of course and psychosocial adjustment equivalent to that obtained in the Collaborative Depression Study--Clinical Branch and other long-term follow-ups. Five year interviews of subjects living within a 250 mile radius will be in person. Furthermore, all records of inpatient or outpatient psychiatric treatment will be obtained and reviewed by the principal investigator who will reconcile these with information from follow-up interviews. Finally, senior clinician will again assign final diagnoses, this time based on all baseline and follow-up information other than neuroendocrine test results.